Measuring Porn on the Web?

Measuring anything on the web is part art and part science. Measuring how much pornography is on the web taxes the art of both. In part, that's because pornography is, at best, loosely defined and subject to local standards. In part, it's because some part of the pornography that's found on the web is not found in the places you might normally measure.

And yet, a U.S. government-commissioned study just identified the percentage of pornography on the web. The study says that about 1 percent of the web contains sexually explicit content.

What is 1 percent of the web? Tough to say. If you do a search on the letter a, the most common letter in the English language, you get 18.5 billion pages. Search on the word porn and you get 88 million pages. Search on porn or xxx and you get 186 million. Roughly 1 percent. If you add the word sex and search for porn or xxx or sex you get something more like 4.4 percent. The first probably undercounts and the second probably overcounts; the truth is probably in the middle. Not a trivial number of pages, but sexually explicit material probably comprises only a small percentage of the entire web.

But that's not the only problem with the study. I can't help but think that if 1 percent of indexed web pages contains sexually explicit content, it's either a very "in your face" 1 percent or that the study might have missed major sources of explicit material.

What sorts of "others" should they factor in? Ads, for one. As I was researching this column for news stories, the ad search results on Google displayed an ever-changing array of ads for pornographic material. Peer-to-peer networks, for another. Sexually explicit images are readily available on many of these networks. Social networking sites like MySpace pages, which also wouldn't show up in an indexed search, might also contain sexually explicit images. A recent survey done at Fresno State that looked at 700 MySpace pages found that 59 percent of the individual pages included risqué/sexual poses, 9 percent included links to pornographic sites, and 6 percent had full frontal nudity of females. 

There are plenty of other reasons for materials not to be indexed as pages: eBay-like sites that compile a set of products for you and are indexed differently. Finally, sites can simply ask that their pages not be indexed by a search engine. I'm sure that others can add to my list.

The study was introduced in court this week as evidence in a complicated ongoing case where the Justice Department is hoping to revive the 1998 Child Online Protection Act. The Act would have required commercial web sites to collect a credit card number or other proof of age before allowing Internet users to view material deemed "harmful to minors." The law was blocked by the Supreme Court in 2004, when it ruled that free speech rights of adults would be hurt and that technologies like filtering software might work better than any legislation.

While the study offers a data point, it needs to be taken within the larger context of how people really see what they see on the Internet. I don't know about you, but if 99 percent of the web is free of sexually explicit material I must be hanging out in the wrong percent.